Generic Automation Examples

n8n examples from GitHub are useful to learn practical workflow patterns before building production automation.

This guide complements:

• content/docs/guides/automation/01-n8n.mdx
• content/docs/guides/automation/02-data-extraction.mdx

1. How to Use This List

• Use template repositories to study workflow structure, not as copy-paste production code.
• Prefer patterns that include retries, validation, and failure handling.
• Keep extraction logic modular so you can swap parser/OCR providers later.
• Record the repository commit or release you used as a reference.

2. Relevant Open-Source Repositories

3. Reusable Patterns You Should Keep

Trigger and Intake

• Webhook, email, Drive, or object storage trigger.
• Normalize input into one event shape early.

{
	"document_id": "doc-001",
	"source": "webhook|email|drive|s3",
	"file_url": "https://...",
	"mime_type": "application/pdf"
}

Route by Document Type

• Use Switch/IF nodes on MIME type, filename, or classifier output.
• Keep one branch per extraction strategy (layout parser, OCR fallback, plain text parser).

Normalize Before Persisting

Use one canonical schema before writing to APIs or databases:

{
	"document_id": "string",
	"checksum": "string",
	"parser_version": "v1",
	"source": "webhook|drive|s3|email",
	"extracted_text": "string",
	"structured": {},
	"confidence": 0.0,
	"extracted_at": "2026-09-27T12:00:00.000Z"
}

Confidence and Human Review

• Auto-process records above threshold.
• Route low-confidence records to manual review.
• Store both original extraction and reviewer-corrected values for auditability.

Dead-Letter and Retry

• Enable retries with backoff on extraction nodes.
• Add an Error Trigger companion workflow.
• Persist failed context to a dead-letter table for triage.

4. Reference Pipeline (Generic)

Document Input
  -> Intake Trigger
  -> Route by Type
  -> Extraction (Parser/OCR)
  -> Normalize Schema
  -> Validation + Confidence
  -> (High) Persist / Index
  -> (Low) Manual Review Queue
  -> Metrics + Alerts

5. 30-Minute Practice Lab

Goal: evaluate one repository pattern and adapt it safely.

1. Pick one template repository from Section 2.
2. Import one workflow into a test n8n instance.
3. Replace external credentials with mock values.
4. Add normalization to the canonical schema from Section 3.
5. Add one retry rule and one dead-letter capture step.
6. Execute 3 tests: valid document, unsupported file type, malformed payload.
7. Capture outcomes in a short decision note: keep, modify, or reject pattern.

Success criteria:

• Workflow runs end-to-end in a test environment.
• Output matches canonical schema.
• Failure case is captured with enough context to debug quickly.

6. Evaluation Checklist

Before adopting a community example, confirm:

• It has clear node-level logic and no hidden assumptions.
• It includes validation and error handling, not only happy-path logic.
• It supports idempotent writes (document_id + checksum).
• It is compatible with your deployment model (local, self-hosted, or cloud).
• It does not expose secrets in workflow JSON exports.

7. Next Step

After selecting a reference pattern:

• Implement the baseline workflow from content/docs/guides/automation/01-n8n.mdx.
• Add the extraction reliability practices from content/docs/guides/automation/02-data-extraction.mdx.